Linear statistical models have been used to describe subjective human judgment. Past studies have focused on the predictive capability of the models, vis-a-vis man himself, the detection and modeling of configural judgment processes, and the use of judgment models as catalysts to conflict resolution. The results have actually not been extensively applied to practical decision situations. The present dissertation is intended to help bridge the gap between the laboratory and operating environments. It applies a concept termed policy capturing to the decision process of an installment loan officer who approves or denies credit on the basis of a written application.